|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
87.228.210.244
I think I may have a partial explanation of the well reported lock-on issue of the AES16 interfacing with an external clock.One of the clocks is 49.152 MHz, which is an exact division by 256 to 192k.
However, the other is 45.168 MHz which when divided by 256 gives 176.4375 kHz. On my system the clock rate for 176.4 k is reported by the card to be 176.4391MHz. Both are more than 200ppm greater than the nominal frequency. This may account for the drop in drop out behaviour and the incomatibilty with the clocking from dCS clcok outputs. In reality, the AES 16 has difficulty both with Big Ben and with clock out on a UA2192 as well.
If true, this is bad design. In the manual, Lynx stresses the importance of narrow plls to kill jitter. If the pll lock-on centre frequency does not correspond accurately to the incoming stream's frequency, then one would expect problems!
On the Lynx forum, there is no acknowledgement of this issue up to a year ago when I gave up looking 'cause I was getting silly answers. Lynx appears to blame this on dCS until then.
Note: I also had a card with aes plugs that won't fit into a standard socket; too big!
Edits: 11/03/09 11/03/09
I noticed about 1 minute for a complete lock, but it always locked for me.
Are you sure it isnt a signal integrity problem? On the Lynx card, it has a 75 ohms to ground terminator at the end and if the clock driver does not have enough drive capacity, it will not work reliably.
Steve N.
100% sure not signal integrity. Fail to lock properly on 3 good clocks with plenty of drive? Actually 4 or 5. One even says it was 10 ppm accurate.
There are quite a few on the Lynx forum, notably dCS users who get no help, just song and dance.
Too bad. It's an expensive card to exhibit this kind of behavior.
Contrast the Fireface400, which you can connect with a hanger wire and the word-clock input will work perfectly, and lock immediately. It all depends on how clever the designer was....
Steve N.
Have you determined which clocks(s) are out of tolerance and by how much?
Last time I worried about such things I was measuring a friend's VHF transceiver. No problem getting accurate frequency measurements using an HP frequency meter slaved to an atomic clock. But I no longer have access to such a hardware lab.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It looks like the 176.4 clock which is 220 ppm out according to Lynx control panel. The 192k clock is 67ppm or so out.
My frequency counter is 5 years old and has not been calibrated since!! I don't have lab support now.
No problem. Any old counter can be calibrated to < <1ppm just using WWV and a pullable reference oscillator. Of course who cares what the absolute frequency is, all that matters is getting the source(s) centered up in the capture range of the PLL. The government isn't totally useless, at least they are willing to give you the time o' day!
Rick
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]